The present invention relates to a multi-gray-scale image display method and an apparatus thereof. More specifically, the present invention relates to a multi-gray-scale image display method and an apparatus thereof that diffuses an adequate amount of error for predetermined upper lines in an image so as to increase the gray-scale image display number.
The use of a digital display device such as a plasma display panel (PDP) in multi-gray-scale image display may degrade the image quality, because multi-gray-scale image display may be beyond the ability of the display device. The gray scale image display number that is subjected to restriction due to physical limitations of the digital display device can be increased by a so-called error diffusion method that uses a spatially averaged gray scale with neighboring pixels.
For example, an 8-bit gray-scale resolution display device actually displays no more than upper eight bits of a 12-bit gray-scale input image signal and leaves the lower four bits that cannot be displayed, as an error component. This error component is multiplied by a predetermined factor and is diffused to the next pixel and its adjacent pixels in the next line, to make the sum of the error components zero over all the pixels as if a 12-bit gray scale image is displayed.
Japanese Patent Application 2000-163005 discloses a conventional multi-gray-scale image display method capable of error diffusion for every piece of digital data even when digital data are input for multiple images.
However, the error diffusion according to the conventional method may lead to missing some of the upper lines. For example, in the error diffusion of the conventional method that is performed equally for all lines of the image, normal error diffusion does not occur in the first line and the gray scale of the first line differs from that of the other lines, because there is no line previous to the first line and hence no error diffusion from the previous line. Similarly, such an abnormal error in the first line is diffused to the second and all the way to about tenth lines.
Particularly, in low-gray-scale image display, about ten upper lines to which an extremely small amount of error is diffused from the previous line are processed as zero by rounding. These about ten upper lines are not displayed at all in the image, thus, the actual size of the image displayed is reduced.
To prevent the line-missing caused by the error diffusion, a method is used to a larger image as shown in FIG. 1, in which the size of the input image is greater by the number of missing lines. However, the method is problematic in that the same image is displayed in different sizes for a display device using error diffusion (e.g., a plasma display panel) and a display device not using error diffusion (e.g., a cathode ray tube).